Image Quality Assessment of Fetal Brain MRI Using Multi‐Instance Deep Learning Methods

• Published in: Journal of magnetic resonance imaging Vol. 54; no. 3; pp. 818 - 829
• Main Authors: Largent, Axel, Kapse, Kushal, Barnett, Scott D., De Asis‐Cruz, Josepheen, Whitehead, Matthew, Murnick, Jonathan, Zhao, Li, Andersen, Nicole, Quistorff, Jessica, Lopez, Catherine, Limperopoulos, Catherine
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.09.2021; Wiley Subscription Services, Inc
• Subjects: Brain; Deep learning; Embedding; fetal brain MRI; Fetuses; Field strength; Gestational age; Image acquisition; Image quality; image quality assessment; Image reconstruction; Inspection; Magnetic resonance imaging; Medical imaging; multi‐instance learning; Neuroimaging; Performance evaluation; Quality assessment; Quality control; Recall; Statistical analysis; Statistical tests; Teaching methods; weakly supervised learning; deep learning; image quality assessment; multi-instance learning; weakly supervised learning; fetal brain MRI
• ISSN: 1053-1807; 1522-2586; 1522-2586
• DOI: 10.1002/jmri.27649

Background Due to random motion of fetuses and maternal respirations, image quality of fetal brain MRIs varies considerably. To address this issue, visual inspection of the images is performed during acquisition phase and after 3D‐reconstruction, and the images are re‐acquired if they are deemed to be of insufficient quality. However, this process is time‐consuming and subjective. Multi‐instance (MI) deep learning methods (DLMs) may perform this task automatically. Purpose To propose an MI count‐based DLM (MI‐CB‐DLM), an MI vote‐based DLM (MI‐VB‐DLM), and an MI feature‐embedding DLM (MI‐FE‐DLM) for automatic assessment of 3D fetal‐brain MR image quality. To quantify influence of fetal gestational age (GA) on DLM performance. Study type Retrospective. Subjects Two hundred and seventy‐one MR exams from 211 fetuses (mean GA ± SD = 30.9 ± 5.5 weeks). Field Strength/Sequence T2‐weighted single‐shot fast spin‐echo acquired at 1.5 T. Assessment The T2‐weighted images were reconstructed in 3D. Then, two fetal neuroradiologists, a clinical neuroscientist, and a fetal MRI technician independently labeled the reconstructed images as 1 or 0 based on image quality (1 = high; 0 = low). These labels were fused and served as ground truth. The proposed DLMs were trained and evaluated using three repeated 10‐fold cross‐validations (training and validation sets of 244 and 27 scans). To quantify GA influence, this variable was included as an input of the DLMs. Statistical Tests DLM performance was evaluated using precision, recall, F‐score, accuracy, and AUC values. Results Precision, recall, F‐score, accuracy, and AUC averaged over the three cross validations were 0.85 ± 0.01, 0.85 ± 0.01, 0.85 ± 0.01, 0.85 ± 0.01, 0.93 ± 0.01, for MI‐CB‐DLM (without GA); 0.75 ± 0.03, 0.75 ± 0.03, 0.75 ± 0.03, 0.75 ± 0.03, 0.81 ± 0.03, for MI‐VB‐DLM (without GA); 0.81 ± 0.01, 0.81 ± 0.01, 0.81 ± 0.01, 0.81 ± 0.01, 0.89 ± 0.01, for MI‐FE‐DLM (without GA); and 0.86 ± 0.01, 0.86 ± 0.01, 0.86 ± 0.01, 0.86 ± 0.01, 0.93 ± 0.01, for MI‐CB‐DLM with GA. Data Conclusion MI‐CB‐DLM performed better than other DLMs. Including GA as an input of MI‐CB‐DLM improved its performance. MI‐CB‐DLM may potentially be used to objectively and rapidly assess fetal MR image quality. Evidence Level 4 Technical Efficacy Stage 3